Closed billdenney closed 2 years ago
This uses the same method as with monolix. I believe that monolix may put this in different places in the file (just like NONMEM), so it sets the compartment property as parsing the code and then later it generates the code in a different location
Unfortunately that means that merge #22 will have to bring this back...
Yeah. That was functionality that I lost in my changes in #22. (Shoot!)
If I follow what you're saying and the code correctly, then these are all moved into the $PK
block in the NONMEM control stream, most of the rest goes into the $DES
block, and a bit goes into the residual error model block. Is that correct?
Overall, it seems to be handled here, is that right? https://github.com/nlmixr2/babelmixr2/blob/3676d28b7689143b90ce2549cc0e9ccf5991f27c/R/nonmemRxUiGetPk.R#L134-L159
And so, we need the .prop
and .pk2
assignment to work correctly (which they won't now), is that right?
In this code, why is
.nonmemSetCmtProperty()
needed rather than directly generating the code for things like lag time and bioavailability?https://github.com/nlmixr2/babelmixr2/blob/5adebcf375d6ef6331940d2d6fe4d16d223c3af4/R/nonmem.R#L547-L614